nit

backend/ee/onyx/db/persona.py

@@ -13,6 +13,7 @@ def make_persona_private(
     persona_id: int,
     user_ids: list[UUID] | None,
     group_ids: list[int] | None,
+    creator_id: UUID | None,
     db_session: Session,
 ) -> None:
     db_session.query(Persona__User).filter(
@@ -26,14 +27,15 @@ def make_persona_private(
         for user_uuid in user_ids:
             db_session.add(Persona__User(persona_id=persona_id, user_id=user_uuid))
 
-            create_notification(
-                user_id=user_uuid,
-                notif_type=NotificationType.PERSONA_SHARED,
-                db_session=db_session,
-                additional_data=PersonaSharedNotificationData(
-                    persona_id=persona_id,
-                ).model_dump(),
-            )
+            if creator_id and creator_id != user_uuid:
+                create_notification(
+                    user_id=user_uuid,
+                    notif_type=NotificationType.PERSONA_SHARED,
+                    db_session=db_session,
+                    additional_data=PersonaSharedNotificationData(
+                        persona_id=persona_id,
+                    ).model_dump(),
+                )
     if group_ids:
         for group_id in group_ids:
             db_session.add(

backend/ee/onyx/db/query_history.py

@@ -132,8 +132,13 @@ def fetch_chat_sessions_eagerly_by_time(
     end: datetime,
     db_session: Session,
     limit: int | None = 500,
+    offset: int | None = 0,
     initial_time: datetime | None = None,
 ) -> list[ChatSession]:
+    """
+    Fetch chunks of ChatSession objects eagerly by time, supporting pagination
+    with limit/offset.
+    """
     time_order: UnaryExpression = desc(ChatSession.time_created)
     message_order: UnaryExpression = asc(ChatMessage.id)
 
@@ -150,6 +155,7 @@ def fetch_chat_sessions_eagerly_by_time(
         .order_by(ChatSession.id, time_order)
         .distinct(ChatSession.id)
         .limit(limit)
+        .offset(offset)
         .subquery()
     )
 
@@ -167,6 +173,4 @@ def fetch_chat_sessions_eagerly_by_time(
         .order_by(time_order, message_order)
     )
 
-    chat_sessions = query.all()
-
-    return chat_sessions
+    return query.all()

backend/ee/onyx/server/query_history/api.py

@@ -1,5 +1,6 @@
 import csv
 import io
+from collections.abc import Generator
 from datetime import datetime
 from datetime import timezone
 from uuid import UUID
@@ -35,6 +36,8 @@ from onyx.server.query_and_chat.models import ChatSessionsResponse
 
 router = APIRouter()
 
+CHUNK_SIZE = 500  # or whatever size is appropriate
+
 
 def fetch_and_process_chat_session_history(
     db_session: Session,
@@ -203,34 +206,61 @@ def get_query_history_as_csv(
     end: datetime | None = None,
     db_session: Session = Depends(get_session),
 ) -> StreamingResponse:
-    complete_chat_session_history = fetch_and_process_chat_session_history(
-        db_session=db_session,
-        start=start or datetime.fromtimestamp(0, tz=timezone.utc),
-        end=end or datetime.now(tz=timezone.utc),
-        feedback_type=None,
-        limit=None,
-    )
+    """Stream the CSV in chunks to avoid timeout and high memory usage."""
+    # Set up the time range defaults
+    start_time = start or datetime.fromtimestamp(0, tz=timezone.utc)
+    end_time = end or datetime.now(tz=timezone.utc)
 
-    question_answer_pairs: list[QuestionAnswerPairSnapshot] = []
-    for chat_session_snapshot in complete_chat_session_history:
-        question_answer_pairs.extend(
-            QuestionAnswerPairSnapshot.from_chat_session_snapshot(chat_session_snapshot)
-        )
+    # Build the streaming generator
+    def generate_csv() -> Generator[str, None, None]:
+        # Prepare CSV writer
+        fieldnames = list(QuestionAnswerPairSnapshot.model_fields.keys())
+        buffer = io.StringIO()
+        writer = csv.DictWriter(buffer, fieldnames=fieldnames)
 
-    # Create an in-memory text stream
-    stream = io.StringIO()
-    writer = csv.DictWriter(
-        stream, fieldnames=list(QuestionAnswerPairSnapshot.model_fields.keys())
-    )
-    writer.writeheader()
-    for row in question_answer_pairs:
-        writer.writerow(row.to_json())
+        writer.writeheader()
+        yield buffer.getvalue()
+        buffer.seek(0)
+        buffer.truncate(0)
 
-    # Reset the stream's position to the start
-    stream.seek(0)
+        # Paginate or chunk the fetching of sessions
+        offset = 0
+        while True:
+            # Example chunk-based fetch from DB
+            chunk_of_chat_sessions = fetch_chat_sessions_eagerly_by_time(
+                start=start_time,
+                end=end_time,
+                db_session=db_session,
+                limit=CHUNK_SIZE,
+                offset=offset,
+            )
+            if not chunk_of_chat_sessions:
+                break  # no more data
+
+            # Convert each chat session to question-answer pairs
+            for chat_session in chunk_of_chat_sessions:
+                chat_snapshot = snapshot_from_chat_session(
+                    chat_session=chat_session,
+                    db_session=db_session,
+                )
+                if chat_snapshot is None:
+                    continue
+
+                # Build Q/A pairs
+                qa_pairs = QuestionAnswerPairSnapshot.from_chat_session_snapshot(
+                    chat_snapshot
+                )
+                for row in qa_pairs:
+                    writer.writerow(row.to_json())
+                    # Yield the CSV row text
+                    yield buffer.getvalue()
+                    buffer.seek(0)
+                    buffer.truncate(0)
+
+            offset += CHUNK_SIZE
 
     return StreamingResponse(
-        iter([stream.getvalue()]),
+        generate_csv(),
         media_type="text/csv",
         headers={"Content-Disposition": "attachment;filename=onyx_query_history.csv"},
     )

backend/onyx/background/celery/apps/beat.py

@@ -156,6 +156,8 @@ class DynamicTenantScheduler(PersistentScheduler):
             logger.info(
                 "_try_updating_schedule: Current schedule is up to date, no changes needed"
             )
+            logger.info("Current schedule is the following:")
+            logger.info(current_schedule)
             return
 
         logger.info(

backend/onyx/background/celery/tasks/indexing/tasks.py

@@ -68,6 +68,8 @@ logger = setup_logger()
 def check_for_indexing(self: Task, *, tenant_id: str | None) -> int | None:
     """a lightweight task used to kick off indexing tasks.
     Occcasionally does some validation of existing state to clear up error conditions"""
+
+    task = f"check_for_indexing start: tenant_id={tenant_id}"
     time_start = time.monotonic()
 
     tasks_created = 0
@@ -86,8 +88,11 @@ def check_for_indexing(self: Task, *, tenant_id: str | None) -> int | None:
 
     # these tasks should never overlap
     if not lock_beat.acquire(blocking=False):
+        task_logger.info(f"{task} - Lock not acquired")
         return None
 
+    task_logger.info(f"{task} - Lock acquired")
+
     try:
         locked = True
 

backend/onyx/background/celery/tasks/shared/tasks.py

@@ -234,6 +234,8 @@ def cloud_beat_task_generator(
     expires: int = BEAT_EXPIRES_DEFAULT,
 ) -> bool | None:
     """a lightweight task used to kick off individual beat tasks per tenant."""
+    task = f"cloud_beat_task_generator start: task_name={task_name}, queue={queue}, priority={priority}"
+
     time_start = time.monotonic()
 
     redis_client = get_redis_client(tenant_id=ONYX_CLOUD_TENANT_ID)
@@ -245,8 +247,10 @@ def cloud_beat_task_generator(
 
     # these tasks should never overlap
     if not lock_beat.acquire(blocking=False):
+        task_logger.info(f"{task} - Lock not acquired")
         return None
 
+    task_logger.info(f"{task} - Lock acquired")
     last_lock_time = time.monotonic()
     tenant_ids: list[str] | list[None] = []
 
@@ -262,6 +266,9 @@ def cloud_beat_task_generator(
             if IGNORED_SYNCING_TENANT_LIST and tenant_id in IGNORED_SYNCING_TENANT_LIST:
                 continue
 
+            task_logger.info(
+                f"{task} - sending task: task_name={task_name}, tenant_id={tenant_id}"
+            )
             self.app.send_task(
                 task_name,
                 kwargs=dict(

backend/onyx/db/persona.py

@@ -163,6 +163,7 @@ def make_persona_private(
     persona_id: int,
     user_ids: list[UUID] | None,
     group_ids: list[int] | None,
+    creator_id: UUID | None,
     db_session: Session,
 ) -> None:
     if user_ids is not None:
@@ -173,14 +174,15 @@ def make_persona_private(
         for user_uuid in user_ids:
             db_session.add(Persona__User(persona_id=persona_id, user_id=user_uuid))
 
-            create_notification(
-                user_id=user_uuid,
-                notif_type=NotificationType.PERSONA_SHARED,
-                db_session=db_session,
-                additional_data=PersonaSharedNotificationData(
-                    persona_id=persona_id,
-                ).model_dump(),
-            )
+            if creator_id and creator_id != user_uuid:
+                create_notification(
+                    user_id=user_uuid,
+                    notif_type=NotificationType.PERSONA_SHARED,
+                    db_session=db_session,
+                    additional_data=PersonaSharedNotificationData(
+                        persona_id=persona_id,
+                    ).model_dump(),
+                )
 
         db_session.commit()
 
@@ -240,6 +242,7 @@ def create_update_persona(
             persona_id=persona.id,
             user_ids=create_persona_request.users,
             group_ids=create_persona_request.groups,
+            creator_id=user.id if user else None,
             db_session=db_session,
         )
 
@@ -275,6 +278,7 @@ def update_persona_shared_users(
         persona_id=persona_id,
         user_ids=user_ids,
         group_ids=None,
+        creator_id=user.id if user else None,
         db_session=db_session,
     )
 

backend/onyx/llm/chat_llm.py

@@ -27,7 +27,6 @@ from langchain_core.prompt_values import PromptValue
 
 from onyx.configs.app_configs import LOG_DANSWER_MODEL_INTERACTIONS
 from onyx.configs.app_configs import MOCK_LLM_RESPONSE
-from onyx.configs.chat_configs import QA_TIMEOUT
 from onyx.configs.model_configs import (
     DISABLE_LITELLM_STREAMING,
 )
@@ -36,7 +35,6 @@ from onyx.configs.model_configs import LITELLM_EXTRA_BODY
 from onyx.llm.interfaces import LLM
 from onyx.llm.interfaces import LLMConfig
 from onyx.llm.interfaces import ToolChoiceOptions
-from onyx.llm.utils import model_is_reasoning_model
 from onyx.server.utils import mask_string
 from onyx.utils.logger import setup_logger
 from onyx.utils.long_term_log import LongTermLogger
@@ -231,15 +229,15 @@ class DefaultMultiLLM(LLM):
     def __init__(
         self,
         api_key: str | None,
+        timeout: int,
         model_provider: str,
         model_name: str,
-        timeout: int | None = None,
         api_base: str | None = None,
         api_version: str | None = None,
         deployment_name: str | None = None,
         max_output_tokens: int | None = None,
         custom_llm_provider: str | None = None,
-        temperature: float | None = None,
+        temperature: float = GEN_AI_TEMPERATURE,
         custom_config: dict[str, str] | None = None,
         extra_headers: dict[str, str] | None = None,
         extra_body: dict | None = LITELLM_EXTRA_BODY,
@@ -247,16 +245,9 @@ class DefaultMultiLLM(LLM):
         long_term_logger: LongTermLogger | None = None,
     ):
         self._timeout = timeout
-        if timeout is None:
-            if model_is_reasoning_model(model_name):
-                self._timeout = QA_TIMEOUT * 10  # Reasoning models are slow
-            else:
-                self._timeout = QA_TIMEOUT
-
-        self._temperature = GEN_AI_TEMPERATURE if temperature is None else temperature
-
         self._model_provider = model_provider
         self._model_version = model_name
+        self._temperature = temperature
         self._api_key = api_key
         self._deployment_name = deployment_name
         self._api_base = api_base

backend/onyx/llm/factory.py

@@ -2,6 +2,7 @@ from typing import Any
 
 from onyx.chat.models import PersonaOverrideConfig
 from onyx.configs.app_configs import DISABLE_GENERATIVE_AI
+from onyx.configs.chat_configs import QA_TIMEOUT
 from onyx.configs.model_configs import GEN_AI_MODEL_FALLBACK_MAX_TOKENS
 from onyx.configs.model_configs import GEN_AI_TEMPERATURE
 from onyx.db.engine import get_session_context_manager
@@ -87,8 +88,8 @@ def get_llms_for_persona(
 
 
 def get_default_llms(
-    timeout: int | None = None,
-    temperature: float | None = None,
+    timeout: int = QA_TIMEOUT,
+    temperature: float = GEN_AI_TEMPERATURE,
     additional_headers: dict[str, str] | None = None,
     long_term_logger: LongTermLogger | None = None,
 ) -> tuple[LLM, LLM]:
@@ -137,7 +138,7 @@ def get_llm(
     api_version: str | None = None,
     custom_config: dict[str, str] | None = None,
     temperature: float | None = None,
-    timeout: int | None = None,
+    timeout: int = QA_TIMEOUT,
     additional_headers: dict[str, str] | None = None,
     long_term_logger: LongTermLogger | None = None,
 ) -> LLM:

backend/onyx/llm/llm_provider_options.py

@@ -29,11 +29,11 @@ OPENAI_PROVIDER_NAME = "openai"
 OPEN_AI_MODEL_NAMES = [
     "o3-mini",
     "o1-mini",
-    "o1",
+    "o1-preview",
+    "o1-2024-12-17",
     "gpt-4",
     "gpt-4o",
     "gpt-4o-mini",
-    "o1-preview",
     "gpt-4-turbo",
     "gpt-4-turbo-preview",
     "gpt-4-1106-preview",

backend/onyx/llm/utils.py

@@ -543,14 +543,3 @@ def model_supports_image_input(model_name: str, model_provider: str) -> bool:
             f"Failed to get model object for {model_provider}/{model_name}"
         )
         return False
-
-
-def model_is_reasoning_model(model_name: str) -> bool:
-    _REASONING_MODEL_NAMES = [
-        "o1",
-        "o1-mini",
-        "o3-mini",
-        "deepseek-reasoner",
-        "deepseek-r1",
-    ]
-    return model_name.lower() in _REASONING_MODEL_NAMES

web/src/app/assistants/mine/AssistantCard.tsx

@@ -147,8 +147,8 @@ const AssistantCard: React.FC<{
               )}
             </div>
             {isOwnedByUser && (
-              <div className="flex ml-2 relative items-center gap-x-2">
-                <Popover>
+              <div className="flex ml-2 relative  items-center gap-x-2">
+                <Popover modal>
                   <PopoverTrigger>
                     <button
                       type="button"
@@ -157,7 +157,9 @@ const AssistantCard: React.FC<{
                       <FiMoreHorizontal size={16} />
                     </button>
                   </PopoverTrigger>
-                  <PopoverContent className={`w-32 z-[10000] p-2`}>
+                  <PopoverContent
+                    className={`w-32 z-[10000] p-2 hover:bg-red-400`}
+                  >
                     <div className="flex flex-col text-sm space-y-1">
                       <button
                         onClick={isOwnedByUser ? handleEdit : undefined}

web/src/app/assistants/mine/AssistantModal.tsx

@@ -8,7 +8,6 @@ import { useUser } from "@/components/user/UserProvider";
 import { FilterIcon } from "lucide-react";
 import { checkUserOwnsAssistant } from "@/lib/assistants/checkOwnership";
 import { Dialog, DialogContent } from "@/components/ui/dialog";
-import { Modal } from "@/components/Modal";
 
 export const AssistantBadgeSelector = ({
   text,
@@ -24,8 +23,8 @@ export const AssistantBadgeSelector = ({
       className={`
         select-none ${
           selected
-            ? "bg-neutral-900 text-white"
-            : "bg-transparent text-neutral-900"
+            ? "bg-background-900 text-white"
+            : "bg-transparent text-text-900"
         } w-12 h-5 text-center px-1 py-0.5 rounded-lg cursor-pointer text-[12px] font-normal leading-[10px] border border-black justify-center items-center gap-1 inline-flex`}
       onClick={toggleFilter}
     >
@@ -117,9 +116,9 @@ export function AssistantModal({
   );
 
   return (
-    <div className="fixed inset-0 bg-neutral-950/80 bg-opacity-50 flex items-center justify-center z-50">
-      <div
-        className="p-0 max-w-4xl  overflow-hidden max-h-[80vh]  w-[95%] bg-background rounded-md shadow-2xl transform transition-all duration-300 ease-in-out relative w-11/12 max-w-4xl pt-10 pb-10 px-10 overflow-hidden flex flex-col"
+    <Dialog open={true} onOpenChange={(open) => !open && hideModal()}>
+      <DialogContent
+        className="p-0 max-h-[80vh] max-w-none w-[95%] bg-background rounded-sm shadow-2xl transform transition-all duration-300 ease-in-out relative w-11/12 max-w-4xl pt-10 pb-10  px-10   flex flex-col   max-w-4xl"
         style={{
           position: "fixed",
           top: "10vh",
@@ -132,11 +131,11 @@ export function AssistantModal({
           <div className="flex flex-col sticky top-0 z-10">
             <div className="flex px-2 justify-between items-center gap-x-2 mb-0">
               <div className="h-12 w-full rounded-lg flex-col justify-center items-start gap-2.5 inline-flex">
-                <div className="h-12 rounded-md w-full shadow-[0px_0px_2px_0px_rgba(0,0,0,0.25)] border border-[#dcdad4] flex items-center px-3">
+                <div className="h-12 rounded-md w-full shadow-[0px_0px_2px_0px_rgba(0,0,0,0.25)] border border-background-300 flex items-center px-3">
                   {!isSearchFocused && (
                     <svg
                       xmlns="http://www.w3.org/2000/svg"
-                      className="h-5 w-5 text-gray-400"
+                      className="h-5 w-5 text-text-400"
                       fill="none"
                       viewBox="0 0 24 24"
                       stroke="currentColor"
@@ -161,15 +160,15 @@ export function AssistantModal({
               </div>
               <button
                 onClick={() => router.push("/assistants/new")}
-                className="h-10 cursor-pointer px-6 py-3 bg-black rounded-md border border-black justify-center items-center gap-2.5 inline-flex"
+                className="h-10 cursor-pointer px-6 py-3 bg-background-800 hover:bg-black rounded-md border border-black justify-center items-center gap-2.5 inline-flex"
               >
-                <div className="text-[#fffcf4] text-lg font-normal leading-normal">
+                <div className="text-text-50 text-lg font-normal leading-normal">
                   Create
                 </div>
               </button>
             </div>
             <div className="px-2 flex py-4 items-center gap-x-2 flex-wrap">
-              <FilterIcon size={16} />
+              <FilterIcon className="text-text-800" size={16} />
               <AssistantBadgeSelector
                 text="Pinned"
                 selected={assistantFilters[AssistantFilter.Pinned]}
@@ -198,15 +197,15 @@ export function AssistantModal({
                 }
               />
             </div>
-            <div className="w-full border-t border-neutral-200" />
+            <div className="w-full border-t border-background-200" />
           </div>
 
           <div className="flex-grow overflow-y-auto">
-            <h2 className="text-2xl font-semibold text-gray-800 mb-2 px-4 py-2">
+            <h2 className="text-2xl font-semibold text-text-800 mb-2 px-4 py-2">
               Featured Assistants
             </h2>
 
-            <div className="w-full px-2 pb-2 grid grid-cols-1 md:grid-cols-2 gap-x-6 gap-y-6">
+            <div className="w-full px-2 pb-10 grid grid-cols-1 md:grid-cols-2 gap-x-6 gap-y-6">
               {featuredAssistants.length > 0 ? (
                 featuredAssistants.map((assistant, index) => (
                   <div key={index}>
@@ -220,7 +219,7 @@ export function AssistantModal({
                   </div>
                 ))
               ) : (
-                <div className="col-span-2 text-center text-gray-500">
+                <div className="col-span-2 text-center text-text-500">
                   No featured assistants match filters
                 </div>
               )}
@@ -228,7 +227,7 @@ export function AssistantModal({
 
             {allAssistants && allAssistants.length > 0 && (
               <>
-                <h2 className="text-2xl font-semibold text-gray-800 mt-4 mb-2 px-4 py-2">
+                <h2 className="text-2xl font-semibold text-text-800 mt-4 mb-2 px-4 py-2">
                   All Assistants
                 </h2>
 
@@ -253,8 +252,8 @@ export function AssistantModal({
             )}
           </div>
         </div>
-      </div>
-    </div>
+      </DialogContent>
+    </Dialog>
   );
 }
 export default AssistantModal;

web/src/components/UserDropdown.tsx

@@ -195,7 +195,7 @@ export function UserDropdown({
             ) : hideUserDropdown ? (
               <DropdownOption
                 onClick={() => router.push("/auth/login")}
-                icon={<UserIcon className="h-5w-5 my-auto " />}
+                icon={<UserIcon className="h-5 w-5 my-auto " />}
                 label="Log In"
               />
             ) : (
@@ -208,14 +208,14 @@ export function UserDropdown({
                       item.svg_logo ? (
                         <div
                           className="
-                        h-4
-                        w-4
-                        my-auto
-                        overflow-hidden
-                        flex
-                        items-center
-                        justify-center
-                      "
+                            h-4
+                            w-4
+                            my-auto
+                            overflow-hidden
+                            flex
+                            items-center
+                            justify-center
+                          "
                           aria-label={item.title}
                         >
                           <svg

web/src/components/popover/Popover.tsx

@@ -62,7 +62,7 @@ export function Popover({
       <RadixPopover.Portal>
         <RadixPopover.Content
           className={`
-            PopoverContent z-[100] 
+            PopoverContent z-50 
             ${contentClassName}
             ${matchWidth ? " PopoverContentMatchWidth" : ""}
           `}

web/src/lib/hooks.ts

@@ -647,11 +647,11 @@ export const useUserGroups = (): {
 
 const MODEL_DISPLAY_NAMES: { [key: string]: string } = {
   // OpenAI models
-  "o1-2025-12-17": "o1 (December 2025)",
-  "o3-mini": "o3 Mini",
-  "o1-mini": "o1 Mini",
-  "o1-preview": "o1 Preview",
-  o1: "o1",
+  "o1-2025-12-17": "O1 (December 2025)",
+  "o3-mini": "O3 Mini",
+  "o1-mini": "O1 Mini",
+  "o1-preview": "O1 Preview",
+  o1: "O1",
   "gpt-4": "GPT 4",
   "gpt-4o": "GPT 4o",
   "gpt-4o-2024-08-06": "GPT 4o (Structured Outputs)",
@@ -753,7 +753,14 @@ export function getDisplayNameForModel(modelName: string): string {
 }
 
 export const defaultModelsByProvider: { [name: string]: string[] } = {
-  openai: ["gpt-4", "gpt-4o", "gpt-4o-mini", "o3-mini", "o1-mini", "o1"],
+  openai: [
+    "gpt-4",
+    "gpt-4o",
+    "gpt-4o-mini",
+    "o3-mini",
+    "o1-mini",
+    "o1-preview",
+  ],
   bedrock: [
     "meta.llama3-1-70b-instruct-v1:0",
     "meta.llama3-1-8b-instruct-v1:0",